This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
The Microdata section was added in response to several use cases provided by various people[1]. Most, if not all, of the people were more interested in incorporating RDFa into HTML5. Rather than incorporate RDFa the HTML5 author unilaterally added a section on a new format, called Microdata. In a previous query to the HTML WG about whether people liked the Microdata section[2], only a few people responded, none form the group who submitted the original use cases for HTML5 metadata. The author of the section has, himself, stated that he doesn't care for it--something reaffirmed with recent IRC discussions involved with testing the Microdata section[3]. None of the original use case submitters were every queried if this met their needs or interests, and none have responded affirmatively to the Microdata section. In fact, most discussion on the section has been negative.[4] In addition, the section contains references to external vocabularies, each of which is maintained by a separate organization, meaning that the mapping between the vocabulary in the Microdata section can, and probably will, become out of date with the external vocabulary. Not only would this not be useful, it could be confusing, and even harmful [5]. There is currently a Raised issue on the Microdata section [5], blocking last call. One way to resolve this issue, and ensure its closure before Last Call is to remove the Microdata section, either into its own spec, or entirely. This would also have the added advantage of reducing what is already an overly large specification. [1] http://lists.w3.org/Archives/Public/public-html/2009May/0207.html [2] http://lists.w3.org/Archives/Public/public-html/2009Jun/0732.html [3] http://krijnhoetmer.nl/irc-logs/whatwg/20090908#l-301 [4] http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2009-May/019601.html [5] http://www.google.com/search?hl=en&lr=&tbo=1&num=50&tbs=qdr%3Ay&q=site%3Alists.whatwg.org+%2Bpredefined&aq=f&oq=&aqi= [5] http://www.w3.org/html/wg/tracker/issues/76
> The Microdata section was added in response to several use cases provided by > various people[1]. Most, if not all, of the people were more interested in > incorporating RDFa into HTML5. This is inaccurate. Most of the use cases actually came from people who are not particularly interested in RDFa. (In fact, it was nigh on impossible to get any use cases from RDFa advocates you yourself even said that RDFa advocates _shouldn't_ give use cases!). > None of the original use case submitters were every queried if this met their > needs or interests This is incorrect. The section was designed in consultation with a number of the people who put forward use cases, including people from large companies such as Yahoo! (though nobody from Google other than myself, who, as you noted, isn't a fan of this entire problem space anyway [1]). > In addition, the section contains references to external vocabularies, each of > which is maintained by a separate organization, meaning that the mapping > between the vocabulary in the Microdata section can, and probably will, become > out of date with the external vocabulary. I disagree with your reasoning (having vocabularies doesn't mean they can't be updated), but in any case this issue has been raised as its own issue separately, so I shall deal with it there. [1] Other people from Google have become interested in Microdata since then, and have offered the resources to perform a usability study to examine some variants of the microdata syntax to see if we can improve it further. I haven't removed the section, since the reasoning for removing it was based on misconceptions.
The emails that drove your decision, which I linked in the bug[1] listed arguments, rather than requirements, but there were some requirements from Dan Brickley, Manu Sporny, Charles McCathieNevile, Julian Reschke, Ben Adida, and Kingsley Idehen. There were a couple of other entries, by an Alex, but I'm not sure who this person is. And the entries by Henri and TJ were along the lines of disagreements with the use cases, not use case entries themselves. The Wiki mentioned in your May email also had entries that were predominately by RDF/RDFa folk. Question: did you ask any of the people who submitted original use cases whether microdata met their requirements? That's not a complicated question, nor is it irrelevant. In fact it is exceptionally relevant. When a query was made whether anyone liked the section, the few people who responded affirmatively were all people who did not provide any use case, as far as I could determine from the documentation you provided. Therefore, microdata did not meet the needs of the use case submitters, and hence has no justification for continued inclusion in the specification. I have no misconceptions about this section. I know exactly why the section was added.
> Question: did you ask any of the people who submitted original use cases > whether microdata met their requirements? That's not a complicated question, > nor is it irrelevant. In fact it is exceptionally relevant. I already answered that yes, the section was designed in consultation with a number of the people who put forward use cases. All the people with whom I discussed use cases indicated that it was satisfactory. Many more have since come forward and have even been using it, e.g.: http://dragnetslegacy.blogspot.com/ > When a query was made whether anyone liked the section, the few people who > responded affirmatively were all people who did not provide any use case, as > far as I could determine from the documentation you provided. > > Therefore, microdata did not meet the needs of the use case submitters, and > hence has no justification for continued inclusion in the specification. Your conclusion does not follow from your premise. Nobody has in fact put forward a use case that microdata fails to handle but which another solution (such as RDFa) does handle. In any case, if you wish this section to be removed from HTML5, please escalate this issue to the chairs. I do not agree with the premise of your argument for removing the section.
It appears this is already covered by ISSUE 76 (<http://www.w3.org/html/wg/tracker/issues/76>)
I will continue to keep this open until Issue 76 has a successful resolution, because one of the HTML WG co-chairs has stated that _this_ is the way to bring about change, not create alternate text. Ian, I listed specific people who provided use cases. Did you ask any of them if Microdata answered their specific concerns and requirements? Listing one site, one site, and giving it as justification for Microdata is frankly, ludicrous. You have no justification for keeping this section, in a specification that is already too large, too complex to be effective as a specification.
(In reply to comment #5) > I will continue to keep this open until Issue 76 has a successful resolution, > because one of the HTML WG co-chairs has stated that _this_ is the way to bring > about change, not create alternate text. Ok. Reassigning to Mike for escalation. > Ian, I listed specific people who provided use cases. Did you ask any of them > if Microdata answered their specific concerns and requirements? Manu Sporny, Julian Reschke, and Ben Adida have all said that Microdata doesn't match their requirements, but I have my doubts that anything that wasn't RDFa would match their requirements. I do not believe I specifically asked the other people you listed (I don't recognise all of their names). Most of the people I requested feedback from directly were specifically not people involved with RDFa or Microformats previously, because I didn't want the input I received to be biased by familiarity with previous solutions (especially since the whole point of the effort was to address the use cases in ways that didn't end up having the same flaws of those other efforts). > Listing one site, one site, and giving it as justification for Microdata is > frankly, ludicrous. I do not feel comfortable listing the names of the people whom I consulted personally. However, I do not feel the need to make any claims to authority to justify the microdata proposal; I feel it is quite capable of justifying its own existence on purely technical merits.
(In reply to comment #6) > (In reply to comment #5) > > I will continue to keep this open until Issue 76 has a successful resolution, > > because one of the HTML WG co-chairs has stated that _this_ is the way to bring > > about change, not create alternate text. > > Ok. Reassigning to Mike for escalation. > Fine > > > Ian, I listed specific people who provided use cases. Did you ask any of them > > if Microdata answered their specific concerns and requirements? > > Manu Sporny, Julian Reschke, and Ben Adida have all said that Microdata doesn't > match their requirements, but I have my doubts that anything that wasn't RDFa > would match their requirements. I do not believe I specifically asked the other > people you listed (I don't recognise all of their names). Most of the people I > requested feedback from directly were specifically not people involved with > RDFa or Microformats previously, because I didn't want the input I received to > be biased by familiarity with previous solutions (especially since the whole > point of the effort was to address the use cases in ways that didn't end up > having the same flaws of those other efforts). > If you didn't get feedback from anyone who provided requirements and use cases, you must realize that your effort was contrary to the interests of the larger community. You let your own unreasonable biases adversely impact on the HTML5 specification. As we have seen, Manu Sporny and others are willing to incorporate RDFa and HTML into a separate but complimentary specification, rather than bloat the already oversized specification. They thought of the greater community, and are working to a compromise. I imagine no one would have a problem if you did the same with Microdata. > > Listing one site, one site, and giving it as justification for Microdata is > > frankly, ludicrous. > > I do not feel comfortable listing the names of the people whom I consulted > personally. However, I do not feel the need to make any claims to authority to > justify the microdata proposal; I feel it is quite capable of justifying its > own existence on purely technical merits. > Anyone not willing to stand up and give their name and their opinion in public is irrelevant. One could just as well reference the Good Fairy and the Mad Hatter for the results to be as valid. So, I will take your answer as no, no one who gave requirements or use cases has responded positively to Microdata.
> If you didn't get feedback from anyone who provided requirements and use cases, > you must realize that your effort was contrary to the interests of the larger > community. I did get feedback from a number of people who provided requirements and use cases. > Anyone not willing to stand up and give their name and their opinion in public > is irrelevant. It's irrelevant how many people are willing to support a feature or not. I'm writing the spec based on technical merit, not appeals to authority, and not based on how many people in the small HTMLWG I can get to support a feature. > So, I will take your answer as no, no one who gave requirements or use cases > has responded positively to Microdata. Then you are either not listening to what I'm saying, or you're calling me a liar. Either way, there's no way to have a productive discussion from that point.
(In reply to comment #8) > > If you didn't get feedback from anyone who provided requirements and use cases, > > you must realize that your effort was contrary to the interests of the larger > > community. > > I did get feedback from a number of people who provided requirements and use > cases. > > > > Anyone not willing to stand up and give their name and their opinion in public > > is irrelevant. > > It's irrelevant how many people are willing to support a feature or not. I'm > writing the spec based on technical merit, not appeals to authority, and not > based on how many people in the small HTMLWG I can get to support a feature. > > > > So, I will take your answer as no, no one who gave requirements or use cases > > has responded positively to Microdata. > > Then you are either not listening to what I'm saying, or you're calling me a > liar. Either way, there's no way to have a productive discussion from that > point. > That is not a mature response. It's also not a response that should be condoned in a W3C woring group. I listed out several people who I know of who gave user requirements and use cases in your initial request. I pulled these names from the documentation you provided. I stated that as far as I know, none of these people have responded positively. You stated others have provided use cases and requirements, and that Microdata fit their needs. I asked, specifically who they are, and what are the use cases, so that we, the global audience, can verify for ourselves how well Microdata meets the requirements. You respond with either I'm not listening, or I'm calling you a liar. Frankly, that's a disturbing response to receive from someone who is the author of a W3C specification. If you're not willing to put the Microdata section to a true test, then it is definitely not good to go for the HTML5 specification, and must be removed. When you're willing to put the section to a rigorous test against all stated use cases (which should become test cases), then we can see about putting it back in.
> I listed out several people who I know of who gave user requirements and use > cases in your initial request. I pulled these names from the documentation you > provided. I stated that as far as I know, none of these people have responded > positively. I confirmed that several have responded negatively. > You stated others have provided use cases and requirements, and that Microdata > fit their needs. I asked, specifically who they are, and what are the use > cases, so that we, the global audience, can verify for ourselves how well > Microdata meets the requirements. I then said that I didn't feel comfortable giving their names. You then said that therefore "no one who gave requirements or use cases has responded positively to Microdata", directly contradicting my statement. > You respond with either I'm not listening, or I'm calling you a liar. When you directly contradict something I'm claiming about my own actions, then those are the only two possibilities I'm aware of. > If you're not willing to put the Microdata section to a true test Not only am I willing to put the Microdata section to a true test, I'm doing it this Thursday and Friday. Google is doing a multiple-person usability study to examine three proposed variants of Microdata to see if we can improve the feature even further. > When you're willing to put the section to a rigorous test against all stated > use cases (which should become test cases), then we can see about putting it > back in. I've already done this. I'm not especially interested in a back-and-forth where you say I'm incompetent, say that what I'm claiming is false, and fail to raise any substantive technical arguments whatsoever. I recommend that you raise this with the chairs.
(In reply to comment #10) > > I listed out several people who I know of who gave user requirements and use > > cases in your initial request. I pulled these names from the documentation you > > provided. I stated that as far as I know, none of these people have responded > > positively. > > I confirmed that several have responded negatively. Yes, I know. > > > > You stated others have provided use cases and requirements, and that Microdata > > fit their needs. I asked, specifically who they are, and what are the use > > cases, so that we, the global audience, can verify for ourselves how well > > Microdata meets the requirements. > > I then said that I didn't feel comfortable giving their names. You then said > that therefore "no one who gave requirements or use cases has responded > positively to Microdata", directly contradicting my statement. > We cannot independently verify with the people directly, nor independently verify their user requirements and use cases. Therefore they cannot be taken into consideration. I suggest you contact them and ask them to make their use cases and requirements public. > > You respond with either I'm not listening, or I'm calling you a liar. > > When you directly contradict something I'm claiming about my own actions, then > those are the only two possibilities I'm aware of. > > I'm saying that we, as the global audience, have a right and responsibility to independently verify your decisions. This requires that we have the same data. I'm asking you to provide this data. If you cannot, then it cannot be considered. > > If you're not willing to put the Microdata section to a true test > > Not only am I willing to put the Microdata section to a true test, I'm doing it > this Thursday and Friday. Google is doing a multiple-person usability study to > examine three proposed variants of Microdata to see if we can improve the > feature even further. > I'm sorry but what a private company does or does not do is irrelevant. We do not have access to the test criteria, Google has shown itself not to be impartial, and we don't have a way of independently verifying the results. Google's data has always been suspect, and will continue to be suspect. > > When you're willing to put the section to a rigorous test against all stated > > use cases (which should become test cases), then we can see about putting it > > back in. > > I've already done this. > No, you have not. > > I'm not especially interested in a back-and-forth where you say I'm > incompetent, say that what I'm claiming is false, and fail to raise any > substantive technical arguments whatsoever. I recommend that you raise this > with the chairs. > I think you're biased, against RDFa, RDF, the W3C, and namespaces, and I think you let your biases influence your decisions. I don't think we have the best HTML5 specification as a result. This item is already a raised issue blocking last call. I will keep this bug open until this issue is resolve, or force to close it by the Chairs.
This discussion is necessary, but also consumes a lot of bandwidth. Therefore I'm asking the chairs to prioritize a decision.
> > I then said that I didn't feel comfortable giving their names. You then said > > that therefore "no one who gave requirements or use cases has responded > > positively to Microdata", directly contradicting my statement. > > We cannot independently verify with the people directly, nor independently > verify their user requirements and use cases. Therefore they cannot be taken > into consideration. I agree. I also don't think they _should_ be taken into consideration. What matters is technical merit, it's not a popularity contest.
(In reply to comment #6) > Ok. Reassigning to Mike for escalation. The issue of whether the Microdata section should be moved to a separate spec is already on the chairs' radar. As Julian notes, we already have an open issue related to this bug -- tracker issue 76, http://www.w3.org/html/wg/tracker/issues/76 So I've added a note there with a link to this bug.
*** Bug 8220 has been marked as a duplicate of this bug. ***
Working Group Decision: Adopt the Change Proposal to Separate Microdata: <http://lists.w3.org/Archives/Public/public-html/2009Dec/0299.html>. Record of decision: <http://lists.w3.org/Archives/Public/public-html/2010Jan/0218.html>.
Checked in as WHATWG revision r4543. Check-in comment: Splitting out the Microdata spec. http://html5.org/tools/web-apps-tracker?from=4542&to=4543